Neville MADDEN <[email protected]> writes:

> Good morning,  I have OSM And+ with the latest update on a Lenovo Idea Pad 
> Pro Tablet.  On a recent trip to India the altimeter seems to be reading 
> about 50m below known altitudes.  (All Indian Railway stations have their 
> Reduced Level (RL) on the station approach sign).  Is there any way I can 
> re-calibrate and adjust the altimeter?

GNSS receivers do not actually have an "altimeter".  They compute a
position, which includes "height above ellipsoid", often written h.
Then, they convert this, using a "geoid model", to "height above geoid",
often written H.  Generally, H corresonds to what people mean when they
say "altitude above mean sea level".  (I say "what they mean" because
really understanding height is tremendously complicated.)

In OsmAnd, there is a data file "World Altitude Corrections", that if
you download, you should see heights/elevations displayed as H.  If you
don't have that, you get h.  For reasons that are unclear to me, this is
marked Android only.

India has pretty high geoid separations -- the difference between geoid
and ellipsoid, written N.  It looks like -68m at Mumbai from a quick
look.  So this could explain your displayed height values.

If anyone understands what happens on iOS, please share.     It's
possible that a model built in to iOS is used.

GNSS receivers typically have builtin models, but also typically they
are extremely poor.  For example around me the real value is roughly
-29m and the models say -33m.   Most programs that use GNSS output (on
android or normal computers) will back out the receivers model.  NMEA
reports H and N, and then one can compute h, which avoids the terrible
model.  You can then apply a better model.  Really - it's that messy.

-- 
You received this message because you are subscribed to the Google Groups 
"OsmAnd" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion visit 
https://groups.google.com/d/msgid/osmand/rmiwm0lxyvr.fsf%40s1.lexort.com.

Reply via email to